*BSD News Article 11592


Return to BSD News archive

Received: by minnie.vk1xwt.ampr.org with NNTP
	id AA1711 ; Tue, 23 Feb 93 14:54:47 EST
Newsgroups: comp.os.386bsd.questions
Path: sserve!manuel.anu.edu.au!munnari.oz.au!news.Hawaii.Edu!ames!elroy.jpl.nasa.gov!swrinde!zaphod.mps.ohio-state.edu!usc!cs.utexas.edu!newsfeed.rice.edu!rice!news.Rice.edu!rich
From: rich@Rice.edu (Richard Murphey)
Subject: Re: SLIP Questions...
In-Reply-To: rkb55989@uxa.cso.uiuc.edu's message of Fri, 19 Feb 1993 11:57:09 GMT
Message-ID: <RICH.93Feb19143729@omicron.Rice.edu>
Sender: news@rice.edu (News)
Reply-To: Rich@Rice.edu
Organization: Department of Electrical and Computer Engineering, Rice
	University
References: <C2p2J9.Aor@news.cso.uiuc.edu>
Date: Fri, 19 Feb 1993 20:37:29 GMT
Lines: 26

In article <C2p2J9.Aor@news.cso.uiuc.edu> rkb55989@uxa.cso.uiuc.edu (Rafal) writes:
   I have a question about the SLIP implementation for 386BSD...

   I just recently got a SLIP line, and I did a little stress-testing
   last night [or earlier this night].  Everything seemed cool, even
   when running X stuff from the school's Sequent and the display set to
   my machine... For a while at least!

   After running for a while, stuff would start to freeze on me, and
   when I tried ping'ing a host on the campus net, i would get lots
   of "Writeto: not enough buffers", or something of that nature 
   errors.  Note that ping was the only app. that even gave me an error
   message... Everything else just froze....

There is a fix for this.  You are loosing an mbuf every time you ping.
I put bundled some of the  patches for slip in
/usr/packages/dialup-slip on ref.tfs.com (140.145.254.251).  The fix
should be either there or in the patch kit. 

   Now if I killed SLATTACH, etc. and re-statred SLIP [w/out re-dialing
   into the SLIP server] everything would work for a while again.  Also
   I would like to note that this "while" couls be as long as an hour
   and a half....

The above contains patches for the slip driver as well which will
improve the reliablilty.  Rich